Span metrics Precision/Recall/F1 are missing (NER)
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 1.3k
- Forks
- 379
- Avg merge
- 3h 30m
- Merged PRs (30d)
- 8
Description
Hi,
Recent updates in tensorflow_text provide capabilities to deal with full tensorflow name entity recognition (NER) using offsets_to_boise_tags and boise_tags_to_offsets. It looks that metrics for ner are still missing... however some previous developments show implementations on that sense (https://github.com/tensorflow/text/pull/262) but it seems it has been moved away right after. Is there any plan to make them officially available into tensorflow_text.metrics?
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reviewing the tensorflow_text.metrics area and the existing NER helpers offsets_to_boise_tags and boise_tags_to_offsets. Use the referenced tensorflow/text pull request 262 as historical context for the earlier implementation. Done means making span-level Precision, Recall, and F1 metrics officially available for full TensorFlow NER.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- tensorflow
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100